Group Number: 10

Names: Chan, C., De Jose, B., Esquillo, V., & Pasia, A.

Final Project: Storytelling with Data

How does the Philippines Compare with fellow ASEAN Nations?

Notebook Primer

Rationale

News headlines often articulate the shortcomings or incompetence of the Philippines in different metrics. But is that really the case? Do Filipinos really have it bad at present? What about before?

It is unfair to compare the Philippines to super power countries in answering this question, and it is for this reason that the study group chose its fellow ASEAN Nations -- Philippines' most immediate neighbors -- as the baseline for comparison.

Areas of Comparison

The areas of comparison for Philippines, and to an effect, each ASEAN Nation, will be: Economic Fitness, Environment, Social, Governance, and their Overall COVID-19 Response

Datasets

The study group will be utilizing four datasets, 2 of which came from World Bank, 1 from WHO, and the last one from Oxford. These are briefly defined below

1. Economic Fitness Dataset

   is the measure of a country’s diversification and ability to produce complex goods on a globally competitive basis

2. Environmental, Social, and Governance Dataset

   has 67 indicators for Environmental, Social and Governance aspects

   For the indicators, the study group chose the following indicators

   Environmental:

   Natural resource depletion is the sum of net forest depletion, energy depletion, and mineral depletion. Net forest depletion is unit resource rents times the excess of roundwood harvest over natural growth. Energy depletion is the ratio of the value of the stock of energy resources to the remaining reserve lifetime. It covers coal, crude oil, and natural gas. Mineral depletion is the ratio of the value of the stock of mineral resources to the remaining reserve lifetime). It covers tin, gold, lead, zinc, iron, copper, nickel, silver, bauxite, and phosphate.

   Social:

   Voice and Accountability captures perceptions of the extent to which a country's citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media. Estimate gives the country's score on the aggregate indicator, in units of a standard normal distribution, i.e. ranging from approximately -2.5 to 2.5.

   Governance:

   Government Effectiveness captures perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies. Estimate gives the country's score on the aggregate indicator, in units of a standard normal distribution, i.e. ranging from approximately -2.5 to 2.5.

3. WHO COVID-19 Global Dataset

   Contains the global data regarding COVID-19

4. Oxford COVID-19 Government Response Tracker

   The Oxford Blavatnik School of Government defines this dataset as

   Governments are taking a wide range of measures to tackle the COVID-19 outbreak. We aim to track and compare worldwide government responses to the coronavirus rigorously and consistently. Systematic information on which measures governments take, and when, can help us understand the responses in a consistent way, aiding efforts to fight the pandemic.

   Our team collects information on common policy responses, scores the stringency of such measures, and aggregates these into a Stringency Index.

Note: All latest data used in the analyses were held only up until December 31, 2020.

Notebook Flow

The study group followed a modular approach as to make it easier to manage the program. Cleaning and processing the datasets required a case-to-case approach, thus, as seen below, each cleaning and processing task was implemented using a function specific to each dataset. In this regard, comments were provided in each function to aid the reader.

With regards to the general flow of the notebook, the block diagram below summarizes the process

Brief insights/observations are interjected after every visualization. Detailed discussions are found after.

Package Dependencies

1. Pandas
2. OpenPyxl
3. Numpy
4. Matplotlib

Load Packages and Data

import pandas as pd
import numpy as np

print('Pandas Version', pd.__version__)
print('Numpy Version', np.__version__)

Pandas Version 1.2.1
Numpy Version 1.19.2

# this cell does indeed take a while to process
# openpyxl is needed for opening the xlsx file for pandas

df_econ = pd.read_csv('data/economic_Data.csv')
df_esg = pd.read_csv('data/ESGData.csv')
df_covid = pd.read_csv('data/WHO-COVID-19-global-data.csv')
df_govres = pd.read_excel('data/OxCGRT_timeseries_all.xlsx', 
                          sheet_name = 'government_response_index')

The function info_prime(df) will be used to gain initial insights into the structure of each dataset.

Data Processing Proper

def info_prime(df):
    
    """General Dataset Information
    
    This functions returns the following: 
    (1) Dataset Summary,  (2) First 5 Rows; and 
    (3) Last 5 rows
    
    Is used for preliminary insight"""
    
    attr_list = [df.info(), df.head(), df.tail()]
    
    i = 0;
    while i < len(attr_list):
        display(attr_list[i])
        i += 1

Economic Fitness Dataset

info_prime(df_econ)

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 298 entries, 0 to 297
Data columns (total 26 columns):
 #   Column          Non-Null Count  Dtype  
---  ------          --------------  -----  
 0   Country Name    298 non-null    object 
 1   Country Code    298 non-null    object 
 2   Indicator Name  298 non-null    object 
 3   Indicator Code  298 non-null    object 
 4   1995            294 non-null    float64
 5   1996            294 non-null    float64
 6   1997            294 non-null    float64
 7   1998            294 non-null    float64
 8   1999            294 non-null    float64
 9   2000            294 non-null    float64
 10  2001            294 non-null    float64
 11  2002            294 non-null    float64
 12  2003            294 non-null    float64
 13  2004            294 non-null    float64
 14  2005            296 non-null    float64
 15  2006            298 non-null    float64
 16  2007            298 non-null    float64
 17  2008            298 non-null    float64
 18  2009            298 non-null    float64
 19  2010            298 non-null    float64
 20  2011            298 non-null    float64
 21  2012            296 non-null    float64
 22  2013            296 non-null    float64
 23  2014            296 non-null    float64
 24  2015            296 non-null    float64
 25  Unnamed: 25     0 non-null      float64
dtypes: float64(22), object(4)
memory usage: 60.7+ KB

None

	Country Name	Country Code	Indicator Name	Indicator Code	1995	1996	1997	1998	1999	2000	...	2007	2008	2009	2010	2011	2012	2013	2014	2015	Unnamed: 25
0	Afghanistan	AFG	Economic Fitness Metric	EF.EFM.OVRL.XD	6.894800e-02	6.449100e-02	2.192500e-01	8.800900e-02	1.412700e-01	2.320000e-01	...	1.454900e-01	2.075800e-01	6.219400e-01	3.112900e-01	1.004800e-01	0.077571	1.015700e-01	5.349700e-02	5.369534e-02	NaN
1	Afghanistan	AFG	Economic Fitness Ranking (1 = high, 149 = low)	EF.EFM.RANK.XD	9.900000e+01	1.060000e+02	8.100000e+01	1.080000e+02	9.900000e+01	8.600000e+01	...	9.900000e+01	9.100000e+01	6.700000e+01	8.300000e+01	1.050000e+02	106.000000	1.010000e+02	1.120000e+02	1.120000e+02	NaN
2	Angola	AGO	Economic Fitness Metric	EF.EFM.OVRL.XD	3.140000e-61	2.210000e-105	3.860000e-113	3.310000e-18	1.600000e-20	5.660000e-41	...	3.950000e-84	4.540000e-70	5.010000e-21	5.030000e-194	4.890000e-253	0.000000	3.810000e-171	3.890000e-233	7.890000e-13	NaN
3	Angola	AGO	Economic Fitness Ranking (1 = high, 149 = low)	EF.EFM.RANK.XD	1.470000e+02	1.470000e+02	1.460000e+02	1.460000e+02	1.460000e+02	1.460000e+02	...	1.470000e+02	1.480000e+02	1.480000e+02	1.480000e+02	1.480000e+02	147.500000	1.480000e+02	1.470000e+02	1.460000e+02	NaN
4	Albania	ALB	Economic Fitness Metric	EF.EFM.OVRL.XD	1.901100e-01	4.401200e-01	3.758900e-01	4.308100e-01	4.499400e-01	5.493900e-01	...	4.816200e-01	8.683600e-01	8.908600e-01	4.264100e-01	4.132700e-01	0.428530	4.238000e-01	3.423200e-01	3.746223e-01	NaN

5 rows × 26 columns

	Country Name	Country Code	Indicator Name	Indicator Code	1995	1996	1997	1998	1999	2000	...	2007	2008	2009	2010	2011	2012	2013	2014	2015	Unnamed: 25
293	South Africa	ZAF	Economic Fitness Ranking (1 = high, 149 = low)	EF.EFM.RANK.XD	44.000000	46.000000	49.00000	51.00000	32.00000	34.00000	...	47.000000	50.000000	46.000000	47.000000	47.000000	46.000000	47.00000	44.000000	45.000000	NaN
294	Zambia	ZMB	Economic Fitness Metric	EF.EFM.OVRL.XD	0.069616	0.027963	0.11811	0.24401	0.25267	0.19920	...	0.077562	0.088999	0.097619	0.071232	0.052598	0.072511	0.11341	0.071836	0.065307	NaN
295	Zambia	ZMB	Economic Fitness Ranking (1 = high, 149 = low)	EF.EFM.RANK.XD	98.000000	122.000000	98.00000	86.00000	82.00000	91.00000	...	114.000000	110.000000	109.000000	113.000000	115.000000	108.000000	100.00000	107.000000	109.000000	NaN
296	Zimbabwe	ZWE	Economic Fitness Metric	EF.EFM.OVRL.XD	0.407790	0.193330	0.15889	0.26544	0.31290	0.43621	...	0.276560	0.273630	0.172350	0.146910	0.159750	0.140470	0.15179	0.163950	0.107817	NaN
297	Zimbabwe	ZWE	Economic Fitness Ranking (1 = high, 149 = low)	EF.EFM.RANK.XD	63.000000	81.000000	91.00000	83.00000	77.00000	72.00000	...	84.000000	85.000000	97.000000	98.000000	95.000000	96.000000	93.00000	93.000000	100.000000	NaN

5 rows × 26 columns

Initialize Data Processing Functions

def asean_econ(df):
    
    """Extracts ASEAN Countries from the Dataset; a function to be used for cleaning.
    
    Is necessary to check if all 10 ASEAN Nations are within the Dataset"""
    
    asean_list = ['Brunei', 'Brunei Darussalam','Cambodia','Indonesia', 'Laos',
                  'Lao PDR', "Lao People's Democratic Republic",'Malaysia', 
                  'Myanmar', 'Philippines','Singapore', 'Thailand',
                  'Vietnam', 'Viet Nam']
    
    #drops the unnecessary columns
    to_drop = ['index', 'Country Code', 'Indicator Code', 'Unnamed: 25']
    df = df[df['Country Name'].isin(asean_list)].reset_index().drop(columns = to_drop)

    print('ASEAN countries extracted and unnecessary columns dropped')
    print('--------------------------------------------------------')
    #Check if unecessary columns remain
    print('Columns Remaining: ', df.columns, sep='\n')
    #Check if all ASEAN Nations were extracted; there should be 10 countries
    print('Countries Extracted: ', df['Country Name'].unique(), sep='\n')
    print('-------------------------------------------------------')
    return df

def clean_econ(df):
    
    """Cleans the Dataset; used as a Higher Order Function"""
    
    #extract the ASEAN Nations from the dataset
    df = asean_econ(df)
    
    #stores the countries into a list; for list comprehensions
    country_list = df['Country Name'].unique().tolist()
    
    df_list = []
    #extracts each country from the dataset and stores it individually into a list
    for country in country_list:
        df_list_elem = df[df['Country Name'].isin([country])]
        df_list.append(df_list_elem)
    
    #formats dataframe into ideal vectors
    #i.e. columns for features, rows for observations, and index for dates
    i = 0
    while i < len(df_list):
        df_list[i] = df_list[i].drop(columns = 'Country Name')
        df_list[i] = df_list[i].set_index('Indicator Name').transpose()
        
        #renames columns and index
        df_list[i] = df_list[i].rename_axis(None, axis = 1)
        
        #name df_list element for future use
        df_list[i].name = country_list[i]
        i += 1
    
    print('Datasets stored into list')
    return df_list

econ_df_list = clean_econ(df_econ)

ASEAN countries extracted and unnecessary columns dropped
--------------------------------------------------------
Columns Remaining: 
Index(['Country Name', 'Indicator Name', '1995', '1996', '1997', '1998',
       '1999', '2000', '2001', '2002', '2003', '2004', '2005', '2006', '2007',
       '2008', '2009', '2010', '2011', '2012', '2013', '2014', '2015'],
      dtype='object')
Countries Extracted: 
['Brunei Darussalam' 'Indonesia' 'Cambodia' 'Lao PDR' 'Myanmar' 'Malaysia'
 'Philippines' 'Singapore' 'Thailand' 'Vietnam']
-------------------------------------------------------
Datasets stored into list

Note

If the countries extracted contained all 10 ASEAN Nations, the processing was a success. These countries were then instantiated as separate datasets from each other, which were then stored into a list. This approach was taken as the notebook revolved around list comprehensions to make our lives easier.

ESG Dataset

info_prime(df_esg)

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 16013 entries, 0 to 16012
Data columns (total 67 columns):
 #   Column          Non-Null Count  Dtype  
---  ------          --------------  -----  
 0   Country Name    16013 non-null  object 
 1   Country Code    16013 non-null  object 
 2   Indicator Name  16013 non-null  object 
 3   Indicator Code  16013 non-null  object 
 4   1960            1352 non-null   float64
 5   1961            1945 non-null   float64
 6   1962            2183 non-null   float64
 7   1963            1958 non-null   float64
 8   1964            1970 non-null   float64
 9   1965            2011 non-null   float64
 10  1966            2012 non-null   float64
 11  1967            2252 non-null   float64
 12  1968            2030 non-null   float64
 13  1969            2042 non-null   float64
 14  1970            3087 non-null   float64
 15  1971            3595 non-null   float64
 16  1972            3837 non-null   float64
 17  1973            3592 non-null   float64
 18  1974            3603 non-null   float64
 19  1975            3941 non-null   float64
 20  1976            3900 non-null   float64
 21  1977            4154 non-null   float64
 22  1978            3912 non-null   float64
 23  1979            3923 non-null   float64
 24  1980            4205 non-null   float64
 25  1981            4234 non-null   float64
 26  1982            4433 non-null   float64
 27  1983            4169 non-null   float64
 28  1984            4207 non-null   float64
 29  1985            4305 non-null   float64
 30  1986            4299 non-null   float64
 31  1987            4602 non-null   float64
 32  1988            4291 non-null   float64
 33  1989            4341 non-null   float64
 34  1990            6595 non-null   float64
 35  1991            6509 non-null   float64
 36  1992            6916 non-null   float64
 37  1993            6696 non-null   float64
 38  1994            6747 non-null   float64
 39  1995            7025 non-null   float64
 40  1996            7961 non-null   float64
 41  1997            7297 non-null   float64
 42  1998            8244 non-null   float64
 43  1999            7218 non-null   float64
 44  2000            9811 non-null   float64
 45  2001            8349 non-null   float64
 46  2002            9908 non-null   float64
 47  2003            9556 non-null   float64
 48  2004            9736 non-null   float64
 49  2005            10066 non-null  float64
 50  2006            9821 non-null   float64
 51  2007            10188 non-null  float64
 52  2008            9776 non-null   float64
 53  2009            9952 non-null   float64
 54  2010            10309 non-null  float64
 55  2011            10064 non-null  float64
 56  2012            10333 non-null  float64
 57  2013            9740 non-null   float64
 58  2014            9854 non-null   float64
 59  2015            9588 non-null   float64
 60  2016            8791 non-null   float64
 61  2017            7418 non-null   float64
 62  2018            6691 non-null   float64
 63  2019            3848 non-null   float64
 64  2020            911 non-null    float64
 65  2050            772 non-null    float64
 66  Unnamed: 66     0 non-null      float64
dtypes: float64(63), object(4)
memory usage: 8.2+ MB

None

	Country Name	Country Code	Indicator Name	Indicator Code	1960	1961	1962	1963	1964	1965	...	2013	2014	2015	2016	2017	2018	2019	2020	2050	Unnamed: 66
0	Arab World	ARB	Access to clean fuels and technologies for coo...	EG.CFT.ACCS.ZS	NaN	NaN	NaN	NaN	NaN	NaN	...	83.533457	83.897596	84.171599	84.510171	NaN	NaN	NaN	NaN	NaN	NaN
1	Arab World	ARB	Access to electricity (% of population)	EG.ELC.ACCS.ZS	NaN	NaN	NaN	NaN	NaN	NaN	...	88.129881	87.275323	88.720097	89.308602	90.283638	89.286856	NaN	NaN	NaN	NaN
2	Arab World	ARB	Adjusted savings: natural resources depletion ...	NY.ADJ.DRES.GN.ZS	NaN	NaN	NaN	NaN	NaN	NaN	...	11.641062	10.437876	6.277652	5.261868	6.385551	8.190277	NaN	NaN	NaN	NaN
3	Arab World	ARB	Adjusted savings: net forest depletion (% of GNI)	NY.ADJ.DFOR.GN.ZS	NaN	NaN	NaN	NaN	NaN	NaN	...	0.055593	0.079402	0.086177	0.076670	0.085885	0.047916	NaN	NaN	NaN	NaN
4	Arab World	ARB	Agricultural land (% of land area)	AG.LND.AGRI.ZS	NaN	27.835643	27.826564	27.845522	27.847925	27.866972	...	36.534503	36.607475	36.624759	36.610850	NaN	NaN	NaN	NaN	NaN	NaN

5 rows × 67 columns

	Country Name	Country Code	Indicator Name	Indicator Code	1960	1961	1962	1963	1964	1965	...	2013	2014	2015	2016	2017	2018	2019	2020	2050	Unnamed: 66
16008	Zimbabwe	ZWE	Strength of legal rights index (0=weak to 12=s...	IC.LGL.CRED.XQ	NaN	NaN	NaN	NaN	NaN	NaN	...	5.00000	5.000000	5.000000	5.000000	5.000000	5.000000	6.000000	NaN	NaN	NaN
16009	Zimbabwe	ZWE	Terrestrial and marine protected areas (% of t...	ER.PTD.TOTL.ZS	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	27.214542	27.214585	27.214585	NaN	NaN	NaN	NaN
16010	Zimbabwe	ZWE	Unemployment, total (% of total labor force) (...	SL.UEM.TOTL.ZS	NaN	NaN	NaN	NaN	NaN	NaN	...	5.39900	5.342000	5.302000	5.253000	5.168000	5.087000	4.954000	4.991	NaN	NaN
16011	Zimbabwe	ZWE	Unmet need for contraception (% of married wom...	SP.UWT.TFRT	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	10.400000	10.400000	NaN	NaN	NaN	NaN	NaN	NaN	NaN
16012	Zimbabwe	ZWE	Voice and Accountability: Estimate	VA.EST	NaN	NaN	NaN	NaN	NaN	NaN	...	-1.38622	-1.256964	-1.165902	-1.177471	-1.194492	-1.123221	-1.141875	NaN	NaN	NaN

5 rows × 67 columns

Initialize Data Processing Functions

As you can see, the indicators of interest were not immediately 'sliced' out. This enabled backtracking much easier should we have changed our minds in the middle of our analysis.

def asean_esg(df):
    
    """Extracts ASEAN Countries from the Dataset; a function to be used for cleaning.
    
    Is necessary to check if all 10 ASEAN Nations are within the Dataset"""
    
    asean_list = ['Brunei', 'Brunei Darussalam','Cambodia','Indonesia', 'Laos',
                  'Lao PDR', "Lao People's Democratic Republic",'Malaysia', 
                  'Myanmar', 'Philippines','Singapore', 'Thailand',
                  'Vietnam', 'Viet Nam']
    
    #drops the unnecessary columns
    to_drop = ['index','Country Code', 'Indicator Code', 'Unnamed: 66']
    df = df[df['Country Name'].isin(asean_list)].reset_index().drop(columns = to_drop)
    
    print('ASEAN countries extracted and unnecessary columns dropped')
    print('--------------------------------------------------------')
    #Check if unecessary columns remain
    print('Columns Remaining: ', df.columns, sep='\n')
    #Check if all ASEAN Nations were extracted; there should be 10 countries
    print('Countries Extracted: ', df['Country Name'].unique(), sep='\n')
    print('-------------------------------------------------------')
    
    return df

def clean_esg(df):
    
    """Cleans the df_esg dataset"""
    
    #extract the ASEAN Nations from the dataset
    df = asean_esg(df)
    
    #stores the countries into a list; for list comprehensions
    country_list = df['Country Name'].unique().tolist()
    
    df_list = []
    #extracts each country from the dataset and stores it individually into a list
    for country in country_list:
        df_list_elem = df[df['Country Name'].isin([country])]
        df_list.append(df_list_elem)
        
    #formats dataframe into ideal vectors
    #i.e. columns for features, rows for observations, and index for dates
    i = 0
    while i < len(df_list):
        to_drop = ['Country Name', '2050']
        df_list[i] = df_list[i].drop(columns = to_drop)
        df_list[i] = df_list[i].set_index('Indicator Name').transpose()
        df_list[i] = df_list[i].rename_axis(None, axis = 1)
        
        #name df_list element for future use
        df_list[i].name = country_list[i]
        i += 1
        
    print('Datasets stored into list')
    return df_list

esg_df_list = clean_esg(df_esg)

ASEAN countries extracted and unnecessary columns dropped
--------------------------------------------------------
Columns Remaining: 
Index(['Country Name', 'Indicator Name', '1960', '1961', '1962', '1963',
       '1964', '1965', '1966', '1967', '1968', '1969', '1970', '1971', '1972',
       '1973', '1974', '1975', '1976', '1977', '1978', '1979', '1980', '1981',
       '1982', '1983', '1984', '1985', '1986', '1987', '1988', '1989', '1990',
       '1991', '1992', '1993', '1994', '1995', '1996', '1997', '1998', '1999',
       '2000', '2001', '2002', '2003', '2004', '2005', '2006', '2007', '2008',
       '2009', '2010', '2011', '2012', '2013', '2014', '2015', '2016', '2017',
       '2018', '2019', '2020', '2050'],
      dtype='object')
Countries Extracted: 
['Brunei Darussalam' 'Cambodia' 'Indonesia' 'Lao PDR' 'Malaysia' 'Myanmar'
 'Philippines' 'Singapore' 'Thailand' 'Vietnam']
-------------------------------------------------------
Datasets stored into list

Note

If the countries extracted contained all 10 ASEAN Nations, the processing was a success. These countries were then instantiated as separate datasets from each other, which were then stored into a list. This approach was taken as the notebook revolved around list comprehensions to make our lives easier.

COVID-19 Datasets

WHO COVID-19 Cases

info_prime(df_covid)

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 92081 entries, 0 to 92080
Data columns (total 8 columns):
 #   Column             Non-Null Count  Dtype 
---  ------             --------------  ----- 
 0   Date_reported      92081 non-null  object
 1   Country_code       91693 non-null  object
 2   Country            92081 non-null  object
 3   WHO_region         92081 non-null  object
 4   New_cases          92081 non-null  int64 
 5   Cumulative_cases   92081 non-null  int64 
 6   New_deaths         92081 non-null  int64 
 7   Cumulative_deaths  92081 non-null  int64 
dtypes: int64(4), object(4)
memory usage: 5.6+ MB

None

	Date_reported	Country_code	Country	WHO_region	New_cases	Cumulative_cases	New_deaths	Cumulative_deaths
0	2020-01-03	AF	Afghanistan	EMRO	0	0	0	0
1	2020-01-04	AF	Afghanistan	EMRO	0	0	0	0
2	2020-01-05	AF	Afghanistan	EMRO	0	0	0	0
3	2020-01-06	AF	Afghanistan	EMRO	0	0	0	0
4	2020-01-07	AF	Afghanistan	EMRO	0	0	0	0

	Date_reported	Country_code	Country	WHO_region	New_cases	Cumulative_cases	New_deaths	Cumulative_deaths
92076	2021-01-20	ZW	Zimbabwe	AFRO	783	28675	52	825
92077	2021-01-21	ZW	Zimbabwe	AFRO	733	29408	54	879
92078	2021-01-22	ZW	Zimbabwe	AFRO	639	30047	38	917
92079	2021-01-23	ZW	Zimbabwe	AFRO	476	30523	45	962
92080	2021-01-24	ZW	Zimbabwe	AFRO	484	31007	12	974

Initialize Data Processing Functions

The data processing function for the WHO COVID-19 Dataset included a normalization code block to normalize each feature. This was conducted to sort of 'scale' each variable such that it will be easier to compare. Furthermore, in case you have not noticed from the past processing functions, the attribute df.name was always created for each dataset so as to make it easier to distinguish which dataframe is which from the returned df_list. The name attribute was also used as a container to make it easier to name titles, and x- or y-labels when it comes to data visualization; this aided in the cumbersome process of manually naming each of the 10 countries for a single plot

def asean_covid(df):
    
    """Extracts ASEAN Countries from the Dataset; a function to be used for cleaning.

    
    Is necessary to check if all 10 ASEAN Nations are within the Dataset"""
    
    #extract data from 2020 only
    df['Date_reported'] = pd.to_datetime(df['Date_reported'], infer_datetime_format=True)
    df = df[(df['Date_reported'].dt.year == 2020)] 
    df.index = df['Date_reported']
    
    asean_list = ['Brunei', 'Brunei Darussalam','Cambodia','Indonesia', 'Laos',
                  'Lao PDR', "Lao People's Democratic Republic",'Malaysia', 
                  'Myanmar', 'Philippines','Singapore', 'Thailand',
                  'Vietnam', 'Viet Nam']
    
    #drops the unnecessary columns
    to_drop = ['Country_code', 'WHO_region', 'Date_reported']
    df = df[df['Country'].isin(asean_list)].drop(columns = to_drop).reset_index()
    
    print('ASEAN countries extracted and unnecessary columns dropped')
    print('--------------------------------------------------------')
    #Check if unecessary columns remain
    print('Columns Remaining: ', df.columns, sep='\n')
    #Check if all ASEAN Nations were extracted; there should be 10 countries
    print('Countries Extracted: ', df['Country'].unique(), sep='\n')
    print('-------------------------------------------------------')
    
    return df

def clean_covid(df):
    
    """Cleans the COVID-19 Dataset"""
    
    #extract the ASEAN Nations from the dataset
    df = asean_covid(df)
    
    #stores the countries into a list; for list comprehensions
    country_list = df['Country'].unique().tolist()
        
    df_list = []
    #extracts each country from the dataset and stores it individually into a list
    for country in country_list:
        df_list_elem = df[df['Country'].isin([country])]
        df_list.append(df_list_elem)    
        
    #formats dataframe into ideal vectors
    #i.e. columns for features, rows for observations, and index for dates
    i = 0
    while i < len(df_list):
        df_list[i] = df_list[i].drop(columns = 'Country')
        df_list[i] = df_list[i].set_index('Date_reported')
        df_list[i] = df_list[i].rename_axis(None, axis = 1)
        i +=1
        
    print('Datasets stored into list')
    
    ## Normalize each variable to per thousands
    
    #Population data retrieved January 25, 2021
    #from https://www.worldometers.info/population/asia/south-eastern-asia/
    pop_brunei = 439902
    pop_cambodia = 16853708
    pop_indo = 275199281
    pop_laos = 7337106
    pop_malay = 32607093
    pop_myanmar = 54619835
    pop_ph = 110429709
    pop_sg = 5876885
    pop_thai = 69900171
    pop_viet = 97338579

    pop_list = [pop_brunei, pop_cambodia, pop_indo,
           pop_laos, pop_malay, pop_myanmar,
           pop_ph, pop_sg, pop_thai, pop_viet]
    j = 0
    while j < len(pop_list):
        pop_list[j] = pop_list[j]/1e3
        j += 1
       
    k = 0 
    while k < len(df_list):
        df_list[k] = df_list[k] / pop_list[k]
        #name df_list element for future use
        df_list[k].name = country_list[k]
        k += 1
    print('Variables normalized to per thousand')
    
    return df_list

covid_df_list = clean_covid(df_covid)

ASEAN countries extracted and unnecessary columns dropped
--------------------------------------------------------
Columns Remaining: 
Index(['Date_reported', 'Country', 'New_cases', 'Cumulative_cases',
       'New_deaths', 'Cumulative_deaths'],
      dtype='object')
Countries Extracted: 
['Brunei Darussalam' 'Cambodia' 'Indonesia'
 "Lao People's Democratic Republic" 'Malaysia' 'Myanmar' 'Philippines'
 'Singapore' 'Thailand' 'Viet Nam']
-------------------------------------------------------
Datasets stored into list
Variables normalized to per thousand

Note

If the countries extracted contained all 10 ASEAN Nations, the processing was a success. These countries were then instantiated as separate datasets from each other, which were then stored into a list. This approach was taken as the notebook revolved around list comprehensions to make our lives easier.

Government Response Index

info_prime(df_govres)

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 185 entries, 0 to 184
Columns: 368 entries, country_code to 31Dec2020
dtypes: float64(366), object(2)
memory usage: 532.0+ KB

None

	country_code	country_name	01Jan2020	02Jan2020	03Jan2020	04Jan2020	05Jan2020	06Jan2020	07Jan2020	08Jan2020	...	22Dec2020	23Dec2020	24Dec2020	25Dec2020	26Dec2020	27Dec2020	28Dec2020	29Dec2020	30Dec2020	31Dec2020
0	ABW	Aruba	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	33.33	33.33	33.33	33.33	33.33	33.33	33.33	33.33	33.33	33.33
1	AFG	Afghanistan	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	22.22	22.22	22.22	22.22	22.22	22.22	22.22	22.22	22.22	22.22
2	AGO	Angola	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	48.89	48.89	48.89	48.89	48.89	48.89	48.89	48.89	48.89	48.89
3	ALB	Albania	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	60.56	60.56	58.33	58.33	58.33	58.33	58.33	58.33	58.33	58.33
4	AND	Andorra	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	63.33	63.33	63.33	63.33	63.33	63.33	63.33	63.33	63.33	63.33

5 rows × 368 columns

	country_code	country_name	01Jan2020	02Jan2020	03Jan2020	04Jan2020	05Jan2020	06Jan2020	07Jan2020	08Jan2020	...	22Dec2020	23Dec2020	24Dec2020	25Dec2020	26Dec2020	27Dec2020	28Dec2020	29Dec2020	30Dec2020	31Dec2020
180	VUT	Vanuatu	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	30.56	30.56	30.56	30.56	30.56	30.56	30.56	30.56	25.56	25.56
181	YEM	Yemen	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	23.89	23.89	23.89	23.89	23.89	23.89	23.89	23.89	23.89	23.89
182	ZAF	South Africa	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	56.67	56.67	56.67	56.67	56.67	56.67	56.67	56.67	56.67	56.67
183	ZMB	Zambia	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	51.11	51.11	51.11	51.11	51.11	51.11	51.11	51.11	51.11	51.11
184	ZWE	Zimbabwe	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	55.56	55.56	55.56	55.56	55.56	55.56	55.56	55.56	55.56	55.56

5 rows × 368 columns

Initialize Data Processing Functions

Nothing special here. This is the usual data processing code block as shown in the block diagram in the notebook primer above.

def asean_govres(df):
    
    """Extracts ASEAN Countries from the Dataset; a function to be used for cleaning.
    
    Is necessary to check if all 10 ASEAN Nations are within the Dataset"""
    
    asean_list = ['Brunei', 'Brunei Darussalam','Cambodia','Indonesia', 'Laos',
                  'Lao PDR', "Lao People's Democratic Republic",'Malaysia', 
                  'Myanmar', 'Philippines','Singapore', 'Thailand',
                  'Vietnam', 'Viet Nam']
    
    #drops the unnecessary columns
    to_drop = ['index', 'country_code']
    df = df[df['country_name'].isin(asean_list)].reset_index().drop(columns = to_drop)
    
    print('ASEAN countries extracted and unnecessary columns dropped')
    print('--------------------------------------------------------')
    #Check if unecessary columns remain
    print('Columns Remaining: ', df.columns, sep='\n')
    #Check if all ASEAN Nations were extracted; there should be 10 countries
    print('Countries Extracted: ', df['country_name'].unique(), sep='\n')
    print('-------------------------------------------------------')
    
    return df

def clean_govres(df):
    
    """Cleans the Oxford Dataset"""
    
    #extract the ASEAN Nations from the dataset
    df = asean_govres(df)
    
    #stores the countries into a list; for list comprehensions
    country_list = df['country_name'].unique().tolist()
        
    df_list = []
    #extracts each country from the dataset and stores it individually into a list
    for country in country_list:
        df_list_elem = df[df['country_name'].isin([country])]
        df_list.append(df_list_elem)    
        
    #formats dataframe into ideal vectors
    #i.e. columns for features, rows for observations, and index for dates
    i = 0
    while i < len(df_list):
        df_list[i] = df_list[i].set_index('country_name').transpose()
        df_list[i].index = pd.to_datetime(df_list[i].index)
        df_list[i] = df_list[i].rename_axis(None, axis = 1)
        
        #name df_list element for future use
        df_list[i].name = country_list[i]
        i +=1
        
    print('Datasets stored into list')
    
    return df_list

govres_df_list = clean_govres(df_govres)

ASEAN countries extracted and unnecessary columns dropped
--------------------------------------------------------
Columns Remaining: 
Index(['country_name', '01Jan2020', '02Jan2020', '03Jan2020', '04Jan2020',
       '05Jan2020', '06Jan2020', '07Jan2020', '08Jan2020', '09Jan2020',
       ...
       '22Dec2020', '23Dec2020', '24Dec2020', '25Dec2020', '26Dec2020',
       '27Dec2020', '28Dec2020', '29Dec2020', '30Dec2020', '31Dec2020'],
      dtype='object', length=367)
Countries Extracted: 
['Brunei' 'Indonesia' 'Cambodia' 'Laos' 'Myanmar' 'Malaysia' 'Philippines'
 'Singapore' 'Thailand' 'Vietnam']
-------------------------------------------------------
Datasets stored into list

Note

If the countries extracted contained all 10 ASEAN Nations, the processing was a success. These countries were then instantiated as separate datasets from each other, which were then stored into a list. This approach was taken as the notebook revolved around list comprehensions to make our lives easier.

Results / Data Visualization Proper

Load Packages

import matplotlib
import matplotlib.pyplot as plt
import matplotlib.lines as mlines

print('Matplotlib Version', matplotlib.__version__)

Matplotlib Version 3.3.2

Initialize Plotting Functions

In generating a boxplot, the study group used the df.boxplot approach instead of using matplotlib. This is because the pandas method made it easier to deal with NaNs. Using the pandas method however, does not enable us to make use of the dictionary that is returned using matplotlib as containers for the legend labels. Thus in the gen_boxplot(df_list, column) function, it can be seen that a proxy_artist was used instead to create legends.

def name(df_list):
    
    """Used to obtain the name for each dataset when generating another one."""
    
    name_list = []
    for df in df_list:
        name_list.append(df.name)
    return name_list

def gen_df(df_list, column):
    
    """Generate DataFrame per column of interest; created to accomodate describe function per country per indicator and 
    the required format in generating a boxplot"""
    
    df_dict = {}
    name_list = name(df_list)
    i = 0
    while i < len(df_list):
        df_dict[name_list[i]] = df_list[i][column].values
        i += 1
        
    df = pd.DataFrame(data = df_dict)
    df.index = df_list[0].index
    pd.to_datetime(df.index)
    
    display(df.describe().T.sort_values(by = 'mean', ascending = False))
            
    return df

def gen_boxplot(df_list, column):
    
    """Generate boxplot for the generated dataframe per indicator; used the boxplot method for pandas because for convenience.
    Pandas df.boxplot approach drops NaN values columnwise"""
    
    df = gen_df(df_list, column)
    df.name = column
    
    #derine text params
    t_wt = 'bold'
    t_size = 'xx-large'
    
    #define artists for legend
    mean_mark = mlines.Line2D([], [], color = 'green', marker  = '^',label = 'Mean', 
                         ls = '')
    median_mark = mlines.Line2D([], [], color = 'orange',label = 'Median', 
                         ls = '-')
    outlier_mark = mlines.Line2D([], [], mec = 'black', mfc = 'white', marker  = 'o',label = 'Outlier', 
                         ls = '')
    
    #plot
    fig, ax = plt.subplots(figsize = (10,5), dpi = 300)
    
    bp = df.boxplot(ax = ax, rot = 20, showmeans = True, showfliers = True,
                    boxprops = dict(linewidth = 1),
                    medianprops = dict(linewidth=1.5),
                    meanprops = dict(marker = '^'))
    
    ax.set_title(df.name, fontsize = t_size, fontweight = t_wt)
    ax.legend(handles = [mean_mark, median_mark, outlier_mark])
    
    
    plt.show()

Descriptive Statistics and Boxplots

For each dataset or indicator, the descriptive statistics and a boxpot is shown. The descriptive statistics is sorted by the mean, from highest to lowest, to make it easier to compare each ASEAN Nation with their performance for each metric

Economic Fitness

col = 'Economic Fitness Metric'

print('-----------------------------------------')
txt = "Descriptive Statistics and Boxplot for '" + col + "' "
print(txt)
gen_boxplot(econ_df_list, col)
print('-----------------------------------------')
print('\n')

-----------------------------------------
Descriptive Statistics and Boxplot for 'Economic Fitness Metric' 

	count	mean	std	min	25%	50%	75%	max
Thailand	21.0	2.182506	0.324422	1.532300e+00	2.051900	2.210700	2.415300	2.673716
Singapore	21.0	1.772365	0.191277	1.465600e+00	1.676200	1.754000	1.847600	2.295572
Malaysia	21.0	1.553873	0.328634	1.052800e+00	1.327900	1.500800	1.861500	2.133432
Indonesia	21.0	1.491691	0.192818	1.030700e+00	1.465800	1.564621	1.595200	1.791400
Vietnam	21.0	1.029285	0.413820	2.602600e-01	0.710280	1.096600	1.373400	1.571400
Philippines	21.0	1.018214	0.340984	6.780300e-01	0.775430	0.902000	1.059400	1.824482
Cambodia	21.0	0.169625	0.090492	2.314300e-02	0.122020	0.171500	0.202000	0.409337
Myanmar	21.0	0.109511	0.029387	3.291100e-02	0.099298	0.110840	0.130440	0.152874
Lao PDR	21.0	0.104160	0.036269	2.541500e-02	0.083057	0.111970	0.124460	0.180170
Brunei Darussalam	21.0	0.011890	0.015106	7.040000e-139	0.000002	0.007098	0.019654	0.056677

-----------------------------------------

Economic Fitness Insights

As observed, Philippines rank somewhere in the middle. Brunei seems to be the lowest because it relies heavily on oil, prompting it not to be economically fit. This means that, if no one buys oil anymore, they will experience economic downfall.

---

Environmental, and Governance

For the indicators, the study group chose the following indicators

Environmental:

Natural resource depletion is the sum of net forest depletion, energy depletion, and mineral depletion. Net forest depletion is unit resource rents times the excess of roundwood harvest over natural growth. Energy depletion is the ratio of the value of the stock of energy resources to the remaining reserve lifetime. It covers coal, crude oil, and natural gas. Mineral depletion is the ratio of the value of the stock of mineral resources to the remaining reserve lifetime). It covers tin, gold, lead, zinc, iron, copper, nickel, silver, bauxite, and phosphate.

Social:

Voice and Accountability captures perceptions of the extent to which a country's citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media. Estimate gives the country's score on the aggregate indicator, in units of a standard normal distribution, i.e. ranging from approximately -2.5 to 2.5.

Governance:

Government Effectiveness captures perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies. Estimate gives the country's score on the aggregate indicator, in units of a standard normal distribution, i.e. ranging from approximately -2.5 to 2.5.

cols = ['Adjusted savings: natural resources depletion (% of GNI)', #Environmental
        'Voice and Accountability: Estimate', #Social
        'Government Effectiveness: Estimate'] #Governance

for col in cols:
    print('-----------------------------------------')
    txt = "Descriptive Statistics and Boxplot for '" + col + "' "
    print(txt)
    gen_boxplot(esg_df_list, col)
    print('-----------------------------------------')
    print('\n')

-----------------------------------------
Descriptive Statistics and Boxplot for 'Adjusted savings: natural resources depletion (% of GNI)' 

	count	mean	std	min	25%	50%	75%	max
Brunei Darussalam	30.0	15.566605	4.174550	8.539486	12.568617	14.905343	18.779019	24.892480
Indonesia	49.0	6.229953	4.086891	0.249513	3.918566	5.313123	7.845093	22.160134
Lao PDR	35.0	5.025913	2.524698	0.799775	3.555244	4.332880	6.351996	10.903614
Vietnam	30.0	4.833473	1.715542	1.752346	3.795227	4.804432	5.849152	8.083903
Malaysia	49.0	4.315210	2.208756	0.161513	3.094410	4.163123	5.558114	9.647225
Myanmar	19.0	2.962865	1.020905	1.193543	2.414638	2.694882	3.529777	5.347948
Cambodia	24.0	1.960822	1.406410	0.459103	1.111217	1.433858	1.907453	5.667436
Philippines	48.0	1.586539	1.016448	0.302752	0.751467	1.448390	2.063541	4.910459
Thailand	48.0	1.456273	0.764432	0.480534	0.866549	1.336729	1.928097	3.312869
Singapore	48.0	0.000121	0.000353	0.000000	0.000000	0.000000	0.000052	0.001779

-----------------------------------------


-----------------------------------------
Descriptive Statistics and Boxplot for 'Voice and Accountability: Estimate' 

	count	mean	std	min	25%	50%	75%	max
Philippines	21.0	0.067571	0.137609	-0.138753	-0.027797	0.030184	0.159266	0.419711
Singapore	21.0	-0.082507	0.168084	-0.387192	-0.183365	-0.098659	0.058619	0.139816
Indonesia	21.0	-0.107171	0.328156	-1.024470	-0.255447	-0.011262	0.131194	0.184858
Malaysia	21.0	-0.373594	0.152082	-0.563193	-0.492930	-0.416986	-0.339679	-0.043430
Thailand	21.0	-0.378744	0.510428	-1.045118	-0.827271	-0.455884	0.126583	0.467803
Brunei Darussalam	21.0	-0.784413	0.167863	-1.098655	-0.901009	-0.776592	-0.650571	-0.505073
Cambodia	21.0	-0.967346	0.141491	-1.219561	-1.105392	-0.931646	-0.899963	-0.740300
Vietnam	21.0	-1.402688	0.102325	-1.538451	-1.474383	-1.403045	-1.366400	-1.087494
Lao PDR	21.0	-1.645094	0.173753	-1.801700	-1.739128	-1.709538	-1.649452	-1.132530
Myanmar	21.0	-1.714552	0.517837	-2.233271	-2.122652	-1.918437	-1.343992	-0.796773

-----------------------------------------


-----------------------------------------
Descriptive Statistics and Boxplot for 'Government Effectiveness: Estimate' 

	count	mean	std	min	25%	50%	75%	max
Singapore	21.0	2.158882	0.142847	1.854913	2.088955	2.199069	2.231474	2.436975
Malaysia	21.0	1.010003	0.165496	0.536283	0.948956	1.021576	1.108519	1.267115
Brunei Darussalam	21.0	0.920303	0.223378	0.243172	0.873299	0.914678	1.059952	1.318309
Thailand	21.0	0.293324	0.089866	0.091475	0.220506	0.315366	0.357056	0.451717
Philippines	21.0	-0.017833	0.127091	-0.313254	-0.103113	0.001555	0.073117	0.193027
Vietnam	21.0	-0.234334	0.187619	-0.580876	-0.382613	-0.247552	-0.070452	0.067400
Indonesia	21.0	-0.254277	0.231833	-0.705336	-0.414211	-0.272911	-0.198749	0.182741
Lao PDR	21.0	-0.773503	0.221791	-1.208072	-0.914389	-0.833849	-0.636936	-0.383220
Cambodia	21.0	-0.825613	0.137479	-1.019837	-0.942989	-0.851850	-0.699455	-0.568605
Myanmar	21.0	-1.348696	0.201490	-1.617900	-1.526451	-1.314760	-1.207594	-0.976280

-----------------------------------------

ESG Data Insights

1. Adjusted savings: natural resources depletion (% of GNI)

   Philippines does not deplete its resouces too much. With regards to the extremes, Brunei is the most natural resource depleting ASEAN Nation, since they rely heavily in drilling Oil. On the other hand, Singapore is the lowest for they rely on imports, not on their natural resources (or the lack thereof).

1. Voice and Accountability: Estimate

   Surprisingly, Philippine Citizens are the allowed the most participation when it comes to selecting its government and the citizen's freedom of speech. This is may be due to the fact that it is a democratic country, with people heaviliy influenced by western ideology. As compared to other ASEAN Nations that are comprised of Monarchies.

2. Government Effectiveness: Estimate

   It was not a surprise that Singapore was the most effective, at least according to the data. Nonetheless, it was comforting to know that Philippines rank somewhere in the middle (even though it lies at the abscissa) when it comes to effective governance among the ASEAN Region.

Overall Government Response to the COVID-19 Pandemic

#generate a dataframe for the cumulative cases per thousand

#please do not mind the descriptive stats below
cumcase_df = gen_df(covid_df_list, 'Cumulative_cases')

	count	mean	std	min	25%	50%	75%	max
Singapore	364.0	5.815655	4.193384	0.0	0.176411	7.523714	9.829995	9.965994
Philippines	364.0	1.320080	1.535238	0.0	0.023114	0.350069	2.827955	4.279030
Indonesia	364.0	0.597594	0.750615	0.0	0.007039	0.212871	1.046701	2.671242
Malaysia	364.0	0.521872	0.768824	0.0	0.093967	0.265019	0.346213	3.388373
Myanmar	364.0	0.333142	0.632572	0.0	0.000348	0.005566	0.249461	2.265477
Brunei Darussalam	364.0	0.256800	0.130026	0.0	0.301203	0.320526	0.331892	0.356898
Thailand	364.0	0.038016	0.022117	0.0	0.027929	0.045486	0.051080	0.095708
Cambodia	364.0	0.010324	0.006970	0.0	0.006512	0.008366	0.016450	0.021716
Viet Nam	364.0	0.006159	0.005117	0.0	0.002350	0.003647	0.011249	0.014958
Lao People's Democratic Republic	364.0	0.002408	0.001596	0.0	0.001363	0.002590	0.003135	0.005588

#Get the latest count (December 31, 2020) of the Cumulative COVID-19 Case per thousands for each country 

print('COVID-19 Cases per Thousands on December 31, 2020')
print('---------------------------------------------------------------')
latest_count = cumcase_df.iloc[-1, :].sort_values(ascending = False)
print(latest_count)
print('---------------------------------------------------------------')

COVID-19 Cases per Thousands on December 31, 2020
---------------------------------------------------------------
Singapore                           9.965994
Philippines                         4.279030
Malaysia                            3.388373
Indonesia                           2.671242
Myanmar                             2.265477
Brunei Darussalam                   0.356898
Thailand                            0.095708
Cambodia                            0.021716
Viet Nam                            0.014958
Lao People's Democratic Republic    0.005588
Name: 2020-12-31 00:00:00, dtype: float64
---------------------------------------------------------------

Actual Visualizations

The visualizations from hereon come in three. The differences in these are the y-limits of each axes. The arrangements of these visualizations follow 3 cases and are stated as follows

Case 1: Uniform Y-Limits for BOTH Cumulative COVID-19 Cases per Thousands and Overall Government Response

Case 2: Uniform Y-Limits for ONLY Overall Government Response

Case 3: Default Y-Limits for both; these were the default scales from matplotlib

The first step was to 'sync' the arrangement of countries of the COVID Dataset list and the Government Response list, since, as stated before, the notebook heavily relies on list comprehension.

govres_df_nlist = [govres_df_list[0], govres_df_list[2], govres_df_list[1], 
                  govres_df_list[3], govres_df_list[5], govres_df_list[4],
                  govres_df_list[6], govres_df_list[7], govres_df_list[8],
                  govres_df_list[9]]

Cumulative Cases per Thousands

Case 1:

Uniform Y-Limits for BOTH Cumulative COVID-19 Cases per Thousands and Overall Government Response

fig = plt.figure(figsize = (20, 25), dpi=300)
i = 0

t_wt = 'bold'
lab_wt = 'semibold'

t_size = 'xx-large'
lab_size = 'large'

row = 5
col = 2
while i < len(covid_df_list):
    ax1 = plt.subplot(row, col, i+1)
    ax1.plot(covid_df_list[i]['Cumulative_cases'], label = 'Cumulative Cases per Thousands',
            color = 'tab:red')
    ax1.set_ylabel('Cumulative Cases per Thousands', color = 'tab:red',
                   fontweight = lab_wt)
    
    ax2 = ax1.twinx()
    ax2.plot(govres_df_nlist[i], label = 'Government Response Score',
            color = 'tab:blue')
    ax2.set_ylabel('Overall Government Response', color='tab:blue',
                   fontweight = lab_wt)
    ax2.set_title(govres_df_list[i].name, fontsize = t_size, fontweight = t_wt)
    
    ax2.set_ylim(bottom = 0, top = 85)
    ax1.set_ylim(bottom = 0, top = 10.5)
    
    i += 1

plt.show()

Case 2:

Uniform Y-Limits for ONLY Overall Government Response

fig = plt.figure(figsize = (20, 25), dpi=300)
i = 0

t_wt = 'bold'
lab_wt = 'semibold'

t_size = 'xx-large'
lab_size = 'large'

row = 5
col = 2
while i < len(covid_df_list):
    ax1 = plt.subplot(row, col, i+1)
    ax1.plot(covid_df_list[i]['Cumulative_cases'], label = 'Cumulative Cases per Thousands',
            color = 'tab:red')
    ax1.set_ylabel('Cumulative Cases per Thousands', color = 'tab:red',
                   fontweight = lab_wt)
    
    ax2 = ax1.twinx()
    ax2.plot(govres_df_nlist[i], label = 'Government Response Score',
            color = 'tab:blue')
    ax2.set_ylabel('Overall Government Response', color='tab:blue',
                   fontweight = lab_wt)
    ax2.set_title(govres_df_list[i].name, fontsize = t_size, fontweight = t_wt)
    
    ax2.set_ylim(bottom = 0, top = 85)
    
    i += 1

plt.show()

Case 3:

Default Y-Limits for both; these were the default scales from matplotlib

fig = plt.figure(figsize = (20, 25), dpi=300)
i = 0

t_wt = 'bold'
lab_wt = 'semibold'

t_size = 'xx-large'
lab_size = 'large'

row = 5
col = 2
while i < len(covid_df_list):
    ax1 = plt.subplot(row, col, i+1)
    ax1.plot(covid_df_list[i]['Cumulative_cases'], label = 'Cumulative Cases per Thousands',
            color = 'tab:red')
    ax1.set_ylabel('Cumulative Cases per Thousands', color = 'tab:red',
                   fontweight = lab_wt)
    
    ax2 = ax1.twinx()
    ax2.plot(govres_df_nlist[i], label = 'Government Response Score',
            color = 'tab:blue')
    ax2.set_ylabel('Overall Government Response', color='tab:blue',
                   fontweight = lab_wt)
    ax2.set_title(govres_df_list[i].name, fontsize = t_size, fontweight = t_wt)
    
    i += 1

plt.show()

COVID-19 Insights

With regards to the initial response to the COVID-19 Pandemic, it can be observed in the plots above that all countries responded immediately after COVID-19 cases were detected in their nations, as illustrated by the sudden spikes in the response scores in the early months.

With regards to effectivity however, with respect to the total cases per thousands, the Philippines come second to the last. Singapore has it the worst. This may be a product of the large population density in Singapore and the Philippines despite other countries having more raw COVID-19 cases.

Daily Cases per Thousands

The visualizations and statistics that follow are not necessarily needed. This is because our study group is initially following Oxford's method of super-imposing the government response scores to cumulative cases. The line plots below serve only its purpose for additional insights regarding the number of "spikes" or "waves" of cases experienced by each ASEAN Country; but the graphs below give insight nonetheless.

print('Descriptive Statistics for Daily Cases per Thousands')
print('-------------------------------------------------------------------------------------------------')
dailycase_df = gen_df(covid_df_list, 'New_cases')
print('-------------------------------------------------------------------------------------------------')

Descriptive Statistics for Daily Cases per Thousands
-------------------------------------------------------------------------------------------------

	count	mean	std	min	25%	50%	75%	max
Singapore	364.0	0.027379	0.041937	0.0	0.001021	0.004594	0.040030	0.242646
Philippines	364.0	0.011756	0.012495	0.0	0.000681	0.008680	0.018600	0.060898
Malaysia	364.0	0.009309	0.015799	0.0	0.000184	0.001135	0.008595	0.071610
Indonesia	364.0	0.007339	0.007725	0.0	0.000523	0.004633	0.013566	0.030411
Myanmar	364.0	0.006224	0.010083	0.0	0.000000	0.000037	0.012294	0.041377
Brunei Darussalam	364.0	0.000980	0.004108	0.0	0.000000	0.000000	0.000000	0.034099
Thailand	364.0	0.000263	0.000784	0.0	0.000014	0.000072	0.000143	0.008240
Cambodia	364.0	0.000060	0.000187	0.0	0.000000	0.000000	0.000059	0.001958
Viet Nam	364.0	0.000041	0.000077	0.0	0.000000	0.000010	0.000041	0.000514
Lao People's Democratic Republic	364.0	0.000015	0.000110	0.0	0.000000	0.000000	0.000000	0.001908

-------------------------------------------------------------------------------------------------

Actual Visualizations

Case 1:

Uniform Y-Limits for BOTH Daily COVID-19 Cases per Thousands and Overall Government Response

fig = plt.figure(figsize = (20, 20), dpi=300)
i = 0

t_wt = 'bold'
lab_wt = 'semibold'

t_size = 'xx-large'
lab_size = 'large'

row = 5
col = 2
while i < len(covid_df_list):
    ax1 = plt.subplot(row, col, i+1)
    ax1.plot(covid_df_list[i]['New_cases'], label = 'Cumulative Cases per Thousands',
            color = 'tab:red')
    ax1.set_ylabel('Daily Cases per Thousands', color = 'tab:red',
                  fontweight = lab_wt)
    
    ax2 = ax1.twinx()
    ax2.plot(govres_df_nlist[i], label = 'Government Response Score',
            color = 'tab:blue')
    ax2.set_ylabel('Overall Government Response Index', color='tab:blue',
                  fontweight = lab_wt)
    ax2.set_title(govres_df_list[i].name, fontsize = t_size, fontweight = t_wt)
    
    ax2.set_ylim(bottom = 0, top = 85)
    ax1.set_ylim(bottom = 0, top = 0.25)
    
    i += 1

plt.show()

Case 2:

Uniform Y-Limits for ONLY Overall Government Response

fig = plt.figure(figsize = (20, 20), dpi=300)
i = 0

t_wt = 'bold'
lab_wt = 'semibold'

t_size = 'xx-large'
lab_size = 'large'

row = 5
col = 2
while i < len(covid_df_list):
    ax1 = plt.subplot(row, col, i+1)
    ax1.plot(covid_df_list[i]['New_cases'], label = 'Cumulative Cases per Thousands',
            color = 'tab:red')
    ax1.set_ylabel('Daily Cases per Thousands', color = 'tab:red',
                  fontweight = lab_wt)
    
    ax2 = ax1.twinx()
    ax2.plot(govres_df_nlist[i], label = 'Government Response Score',
            color = 'tab:blue')
    ax2.set_ylabel('Overall Government Response Index', color='tab:blue',
                  fontweight = lab_wt)
    ax2.set_title(govres_df_list[i].name, fontsize = t_size, fontweight = t_wt)
    
    ax2.set_ylim(bottom = 0, top = 85)
    
    i += 1

plt.show()

Case 3:

Default Y-Limits for both; these were the default scales from matplotlib

fig = plt.figure(figsize = (20, 20), dpi=300)
i = 0

t_wt = 'bold'
lab_wt = 'semibold'

t_size = 'xx-large'
lab_size = 'large'

row = 5
col = 2
while i < len(covid_df_list):
    ax1 = plt.subplot(row, col, i+1)
    ax1.plot(covid_df_list[i]['New_cases'], label = 'Cumulative Cases per Thousands',
            color = 'tab:red')
    ax1.set_ylabel('Daily Cases per Thousands', color = 'tab:red',
                  fontweight = lab_wt)
    
    ax2 = ax1.twinx()
    ax2.plot(govres_df_nlist[i], label = 'Government Response Score',
            color = 'tab:blue')
    ax2.set_ylabel('Overall Government Response Index', color='tab:blue',
                  fontweight = lab_wt)
    ax2.set_title(govres_df_list[i].name, fontsize = t_size, fontweight = t_wt)
    
    i += 1

plt.show()

Discussion of Results / Visualizations

Economic Fitness

As defined, Economic Fitness (EF) is a measure of a country's ability to produce complex goods on a globally competitve basis. Achieving a high value of EF indicates that a country has enough capabilities to product a wide-range of products, aquire more predictable long-term growth, and secure good competitive position against other countries. Meanwhile, acquiring a low value of EF signifies that the country is most likely to suffer from poverty, with low capabilities and unpredictable growth.

As we refer to the visualization on the Economic Fitness of ASEAN countries, one could perceive that *Brunei, Laos, Cambodia, and Myanmar* have the least measure of EF, which, by definition, indicates that such countries are under extreme poverty and have difficulty with competing globally in producing complex goods. It could be viewed that *Brunei*'s boxplot measure is at an average of 0.011890, with highest EF measure reaching up to only 0.056677, lowest at 0. However, it is likely that the Brunei exhibits an extremely low measure due to its extreme reliance on oil, prompting it to not be economically fit. In a situation where oil is oil isn't as important as it is now, the country is sure to experience economic downfall. *Laos* measures an average EF value of 0.104160, achieving at most 0.180170 throughout the years, a possible effect of the country's underdeveloped financial sector and delayed private-sector growth (Index of Economic Freedom, n.d). *Cambodia* has garnerned an average EF measure of 0.169625 and has ranged from 0.023143 to 0.409337. Although, we must note that the highest EF value attained serves as an outlier to the dataset of Cambodia. Lastly, *Myanmar* has generated a 0.109511 EF average, with minimum and maximum measures ranging from 0.032911 to 0.152874. Furthermore, remaining ASEAN countries such as *Indonesia, Malaysia, Philippines, Singapore, Thailand, and Vietnam* have acquired an average EF of atleast 1.0, with *Thailand* having the highest average EF measure of 2.182506, such a measure was attainable due to the country's GDP growth and continuously expanding economic freedom, as it shifts from a low-income to an upper-income country (Index of Economic Freedom, n.d.). This is followed by *Singapore* whose EF measure ranges from 1.465600 to 2.295572, with an average of 1.772365. Subsequently, *Malaysia*, obtained an EF measure of 1.553873, *Indonesia* with 1.491691, *Vietnam* acquiring a measure of 1.029285, and lastly, *Philippines* with an EF of 1.018214. The aforementioned countries acquiring atleast a 1.0 measure signfies that they are performing better in terms of producing a vast-range of products, with foreseeable growth.

Additionally, the data visualization of the EF measures shows that *Indonesia, Philippines, Cambodia, and Singapore* have outliers in their datasets, indicating that there were instances throughout the years 1995 to 2015 where EF measures may have surged or plummeted, possibly brought about by the country's competitiveness and capabilities.

Ranking all the ASEAN countries based on its average EF measure, *Thailand* would rank first, followed by *Singapore, Malaysira, Indonesia, Vietnam, Philippines, Cambodia, Myanmar, Laos* with *Brunei* ranking last as it has the least EF measure from the years 1995 to 2015. Given that the dataset wasn't able to capture the Economic Fitness measure from five years ago (2016 to 2020), it is highly likely that the EF measures of the mentioned countries, as well as their ranks, are subject to change.

Environmental, Social and Governance

1. Environmental

   Brunei is one of the countries with the most depleted resources with the country being one of the main resources for oil. This highly affected its rank giving a high score in natural resource depletion. Philippines being rich in resources does not deplete its resources as much.

2. Social

   With democratic countries such as Philippines, the voices of the citizens are highly valued thus resulting to high social scores as compared to other countries. Freedom is respected in the country which gives the people the power to choose, express, and associate.

3. Governance

   With effectiveness of the government, Singapore is the leading among the countries. With Singapore’s highly developed economy this does not come as a surprise. Additionally, Philippines as one of the developing countries did not come far behind as it is still on its way to becoming one of the greatest.

COVID-19 Crisis Response

In terms of case 1 for cumulative cases, if one looks closely at the graphs, notice the spike in government response after just a few days of the first case per country. We can infer that each country responded immediately which indicates a positive sign. Whether these were through stay-at-home orders, lockdowns, border closures, and/or other financial government response tools, these were common to most countries in curbing the spread of COVID-19.

Another observation is that compared to other ASEAN countries, Singapore has noticeably peaked in terms of their cumulative cases, and Philippines follows as the second worse. But as we looked at other references, Singapore’s average response in managing the pandemic actually ranks comparatively high. For that reason, we interjected the population density and found out that the reason why Singapore has the highest covid-19 cases among ASEAN nations, despite being normalized to “per thousands”, is because their population density is quite high. In fact, it has the second greatest population density in the world. So because the people are compacted, the virus is most likely to spread easily. Then again that is not to undermine the level of response the government puts in. For Cambodia, it is also possible that its location may have played a role in its success for curbing the spread as the country is surrounded by Vietnam, Laos, and Thailand, all of which controlled the virus relatively well for most of 2020. In the latter part of 2020, however, we can observe that the line increases for Cambodia and other countries which means that there could have been an outbreak of imported cases which called for the increase in government response.

The case 2 for cumulative cases just gives it another perspective for us, but we can see based on the y-limits for cumulative cases how minute their intervals are like Laos, Indonesia, and Vietnam as compared to the other countries. Nevertheless, we dont think that the government should be complacent in responding to these cases. Overall, the level of government response is able to keep up with the cumulative cases–and in some cases, the government had even outdone themselves.

With the population density in mind, the daily cases per thousands was added. For the first case, it is possible that in some countries, the spread may have increased but the government was able to curb the spread right away (e.g. Brunei). If we turn our attention again to Singapore, we can see that there is a sharp increase in the number of cases in May. This could be linked to overseas workers who are required to be quarantined in dormitories. So although we may see a high government response during that peak, it is possible that they may have strayed their attention away from overseas workers and merely focused on the natives. This would also be a similar case for the Philippines in terms of our OFWs because once they are directed to a quarantine facility, a lot of them stay in one room so there is no proper segregtion and social distancing. In short, we infer that there may be a poor response directed to overseas workers which calls for more attention.

--END--